much wider (between
(L/B=2.69) the
and
shallower
(B/T=3.74)
compared to the Series 60 hull with L/B=7.5 and B/T=2.5. The flow on the downstream side of the escort tug
waterline and the keel) was
proportionally faster than the flow on the downstream side of the Series 60 hull, while the flow over the bottom was approximately the same. As a result, there was less of a shear force gradient on the tug and so when the vortex forms from the downstream bilge corner it is not as strong as the vortex on the Series 60 model.
8.
RECOMMENDATIONS FOR FURTHER STUDY
There are some improvements that could be made to the CFD mesh that might improve the level of prediction of the forces and flow patterns. The first major refinement would be to include the free surface waves generated by the hull. This was ignored from the current meshes on the basis that the effect of the free surface on the forces measured in the model experiments was seen to be small. The free surface of the water will distort and may affect the flow patterns close to the surface. This effect will become more noticeable as yaw angles and flow speeds increase.
Another refinement would be to make the mesh elements smaller in key areas of the flow. The most likely areas for refinement are where vortices are generated in the flow. The most noticeable vortices observed in the PIV experiments were around the downstream bilge for the hull without the fin, and the large vortex generated by the fin when it was fitted. A refined mesh could be generated and the results compared with the single measurement window PIV data (Molyneux et al. 2007), instead of the coarser data spacing that was used for the complete data set.
9. CONCLUSIONS
A commercial RANS Computational Fluid Dynamics (CFD) code was used to predict the forces generated by an escort tug hull, and the same hull fitted with a low aspect ratio fin, over the typical operating range of yaw angles, from 10 to 60 degrees. Two types of mesh were used. One type was a tetrahedral mesh, consisting of elements with four, three sided faces. The other type was a hexahedral mesh, consisting of elements made of six four sided faces. The most accurate force predictions were obtained
using the mesh made entirely of
hexahedral elements. This mesh gave force predictions that on average were within 5-6 % of measured values for the same flow conditions, and never exceeded 10%. The number of elements for the hexahedral mesh was less than one half of the number in the tetrahedral mesh, which resulted in a faster solution time.
The flow patterns around the hull predicted by both meshes at 45 degrees yaw were compared to PIV measurements taken at two planes around the hull. A subjective comparison of the results indicated that the hexahedral mesh gave slightly better predictions of the flow patterns, especially for the flow conditions across the bottom of the hull. A numerical analysis comparing the two meshes over the complete measurement region indicated that the differences were very localized and numerically very small. The average difference between the measured and predicted in-plane flow velocity vector magnitudes was between 8 and 10 per cent.
When the data for forces and flow patterns were combined, the best approach for creating a CFD simulation of an escort tug operating at a large yaw angle was to use a hexahedral mesh. Earlier CFD studies on the Series 60 (Molyneux & Bose, 2007) indicated that neither meshing approach had a significant advantage, but this conclusion was based principally on flow data and only included force measurements at 10 degrees of yaw. The different shape of the hull for the escort tug may have an effect on the accuracy of the predictions for different meshes, since this hull was wide and shallow with a high degree of curvature, whereas the Series 60 was relatively narrow with very sharp waterlines in the bow and stern. This narrow entrance angle for the Series 60 hull creates a vortex, which is generated from the keel at the bow and moves towards the downstream side of the hull. For the escort tug, the fin is the only thing creating a large vortex, and the flow remains attached to the hull, even on the down stream side.
10. ACKNOWLEDGEMENTS
The work described in this paper would not have been possible without the help and support of many people, which is gratefully acknowledged:
Mr. Robert Allan, President of Robert Allan Ltd., Vancouver, British Columbia, for permission to use the model of the escort tug in the PIV experiments and for the use of the experimental data on the force components to compare with the CFD simulations. His enthusiasm for the subject of tug design and encouragement for me to carry out this work is also very much appreciated.
The Canada Foundation for Innovation and the
Newfoundland and Labrador Department of Innovation, Trade and Rural Development for financial support of the purchase of the PIV system used for the escort tug experiments.
Ms. Jie Xu, research laboratory coordinator in the
Department of Ocean and Naval Architectural Engineering, for her leadership, continuous dedication and attention to detail that was necessary for making successful PIV experiments.
©2008: Royal Institution of Naval Architects
B-59
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68